*** REPLICATION FILE FOR PSRM PAPER. GOOCH & VAVRECK 3/4/2016.  
** This file produces results for the discussion in the paper on recency.


*Open the datasetclear
use "GoochVavreck_PSRM3.dta"

	
	
******************************PRIMACY/RECENCY TESTS (in text, no table)
** Does distribution change based on the ORDER of the choices? (Here, you're looking at the chi-square test for each cross-tab)
set more off

foreach v in retro taxes abort gaymar immi choice08 match_mr match_jh partyid bible moby {
	tab `v' `v'rand, col chi
	bysort treat: tab `v' `v'rand, col chi
	}

** Same as above but for the racial resentment questions separately
foreach v in rr_1 rr_2 rr_3 rr_4 {
	tab `v' `v'rand, col chi
	bysort treat: tab `v' `v'rand, col chi
	}
	
	
